1. CRESST Cryogenic Rare Event Search with Superconducting Thermometers Max-Planck-Institut für Physik University of Oxford Technische Universität München Laboratori Nazionali del Gran Sasso Universität Tübingen Outline: CRESST-II detector concept Quenching factor measurements First results with two CRESST-II prototype detector modules (setup without neutron shield) Status of Upgrade Franz Pröbst MPI für Physik CRESST

2. CRESST-II Detector Concept Discrimination of nuclear recoils from radioactive  +  backgrounds (electron recoils) by simultaneous measurement of phonons and scintillation light Separate calorimeter as light detector light reflector W-thermometer 300 g scintillating CaWO 4 crystal proof of principle Energy in light channel keV ee ] Energy in phonon channel [keV]

3. 300 g Prototype Detector Module CRESST-II: 33 detector modules  66 readout channels

4. Recoil energy spectrum in CaWO 4 expected from neutrons at Gran Sasso (no neutron shield) Contribution of W recoils negligible above 12 keV   A 2 for WIMPs with spin independent interaction WIMPs dominantly scatter on W (A=184) nuclei Monte Carlo simulation dry concrete

5. Quenching factor measurement with TOF PMT UV Laser W, O, Ca ions CaWO 4 crystal PTFE reflector collimator Ion source UV Laser desorbs singly or doubly charged ions from almost any material. Acceleration to 18 keV (or 32 keV for double charged) Mount CaWO 4 crystal on PMT at end of flight tube and count single photons after arrival of ion. Adjust laser intensity such that more than 1 ion arriving per laser shot is negligible. Deflection plate for ion type selection target

6. Single photon counting after arrival of ion

7. Quenching factor measurement with TOF Light curve of scintillatorPhoton counts per laser shot Comparision with photon counts per 6 keV x-ray yields quenching factor

8. Quenching Factors for various nuclei in CaWO 4 High value of Q=40 for tungsten  very little light for recoils <40 keV at 300 K at 7 mK Discriminate W recoils (WIMPS) from O/Ca recoils of neutrons

9. Data taken with two 300 g prototype modules (3.2 ± 0.5) counts/day  : 1.0 keV @ 46.5 keV:  : 6.7 keV @ 2.3 MeV Very stable response over a period of 40 days Energy resolution of phonon channelPulse height of heater pulses 1.0 keV FWHM

10. 90% of nuclear recoils with quenching factor expected for neutron induced recoils below this line. Rate=0.87  0.22 /kg/day compatible with expec- ted neutron rate. Low Energy Event Distribution no neutron shield 10.72 kg days 90% of nuclear recoils with quenching factor 40 (W) below this line. No events in 12 to 40 keV range

11. Upper limit for spin indepentend WIMP nucleon cross section 6.17 kg days W exposure

12. Stability of exlusion limit Variation of threshold Variation of quenching factor Including the 2 events at 10.5 and 11.2 keV has practically no effect on result. Variation of Q well beyond uncertainties has only small effect:  =1.6x10 -6  2.3x10 -6 pb

13. Detector Performance at high Energies  Excellent linearity and energy resolution at high energies  Perfect discrimination of  +  from  ’s  Identification of alpha emitters  peaks  and 

14. 232 Th calibration: Linearity of detector response at high energies

15. Identification of  -Emitters  Same light for extern and intern 210 Po  no surface degradation  Relatively low alpha rates: total ~ 2mBq/kg. 238 U ~2 10 -12 g/g ??

16. 147 Sm 152 Gd 144 Nd 180 W

17. exposure28.62 kg days Half life T 1/2 = (1.8±0.2) x 10 18 years Energy Q = (2516.4±1.1 (stat.)±1.2(sys.)) keV Phys. Rev. C 70 (2004) 64606 Half-life for the α-decay of 180 W

18. Status of Upgrade Read out and electronics: 66 SQUIDs for 33 detector modules (10 kg target) and DAQ ready Neutron shield: 50 cm polyethylen (installation complete) Muon veto: 20 plastic scintillator pannels outside Cu/Pb shield and radon box. Analog fiber transmission through Faraday cage (ready) Detector integration in cold box and wiring (entering fabrication stage)

19. Detector integration in existing cold box Mounting of 33 detector modules at end of cold finger Detectors indivi- dually exchangeable Spring suspension for vibration decoupling from cryostat.

20. Conclusion Two CRESST II prototype detectors have been operated for two months. Discrimination threshold (  ) well below 10 keV Type of recoiling nucleus identified above 12 keV Upgrade to 66 readout channels (10 kg target), installation ofneutron shield and muon veto almost complete. Restart in summer/autumn 2005 CRESST-II is aiming for a sensitvity of  =10 -8 pb